promptfoo-framework-nist-ai-rmflisted

# NIST AI Risk Management Framework The NIST AI RMF is a voluntary framework for managing AI risks. Promptfoo focuses on the **Measure** function — testing and evaluating AI systems against specific risk metrics. Use `nist:ai:measure` for comprehensive testing or target specific measures (2.4 safety, 2.7 security, 2.8 privacy, 2.11 fairness). ## Use this skill when - Testing AI systems against NIST AI RMF Measure categories - Demonstrating NIST AI RMF compliance for federal AI guidelines - Measuring safety risks (2.4), security and resilience (2.7), privacy (2.8, 2.10), fairness and bias (2.11) - Testing misuse and abuse potential (2.6) - Setting up continuous testing to meet "regularly assessed" requirements ## Do not use this skill when - Testing OWASP LLM Top 10 specifically → use `promptfoo-framework-owasp-llm` - Testing GDPR data protection → use `promptfoo-framework-gdpr` - Testing EU AI Act compliance → use `promptfoo-framework-eu-ai-act` - Testing DoD AI ethics → use `promptfoo-framework-dod-ai-ethics` - Testing general LLM app security without a framework → use `promptfoo-redteam-llm` ## 🛡️ Edge cases (mandatory handling) - **Automated testing covers Measure only** — Govern, Map, and Manage functions need organizational processes beyond promptfoo. - **MEASURE 2.12 (environmental impact)** — requires infrastructure monitoring, not testable via red teaming. - **MEASURE 3.3 (stakeholder feedback)** — requires organizational procedures for user feedback. - **MEASURE